Reading disability (RD) is characterized by unexpected reading difficulty in children and adults who otherwise have the intelligence and instructional opportunity necessary for accurate and fluent reading. Worldwide, the reported prevalence ranges from 5 to17%. In 1995-96 US schools spent more than $30billion on remediation, mostly for RD. Yet, RD is frequently unrecognized, leading to academic under-achievement with detrimental social and economic consequences. Intervention programs work, but are most effective when RD is diagnosed at an early age. Studies of twins show that the heritability is 44-77%. Four RD genes have emerged from chromosomal loci initially identified by genetic linkage: DYX1C1 (15q), ROBO1 (3q), KIAA0319 (6p22), and DCDC2 (6p22). Evidence for genetic association for KIAA0319 and DCDC2 is particularly strong. The KIAA0319 association has been independently replicated in both the US and the UK, while the DCDC2 association has been independently replicated in the US, Germany, and Italy. All four RD genes have potent effects on neuronal migration. We hypothesize that the risk of RD that is attributable to these specific genes is substantial. To test this hypothesis, we will determine the individual and population attributable risks for RD conferred by all single or combinations of alleles from these genes in a sample of 10,233 randomly selected children from the Avon Longitudinal Study of Parents and Children (ALSPAC). To determine the non-genetic factors that contribute to RD, we will first model the effects of school, home and individual characteristics and their interactions on a variety of cognitive, academic, and behavioral measures at different ages in order to identify significant covariates for genetic studies. To determine the risk of RD attributable to these genes, we will genotype all the known RD-alleles in the ALSPAC, including KIAA0319, DCDC2, intervening 6p22 markers, BV677278 alleles, ROBO1, and DYX1C1. We will then test for association with single alleles and reconstructed haplotypes conditioned by the covariate modeling in AIM 1, define the attributable risks of RD for any child and for the population associated with any single RD-allele and with all combinations of alleles. We will also assess gene-gene and gene-environment interactions, and epigenetic effects. Finally, to validate these results, we will attempt to replicate our positive findings in a random, sub- sample of 700 ALSPAC trios, in which we will correct for population admixture as well as assess parent-of- origin effects that can arise by epigenetic effects. We expect these studies will confirm and quantify the substantial risk of RD conferred by these genes which should stimulate further studies to lay the groundwork for implementing genetic screening programs for RD. Such screening programs could lead to early identification of those at increased risk, as well as to early interventions for those that are affected, thereby decreasing the considerable morbidity caused by RD. PUBLIC HEALTH RELEVANCE: The goal of these studies is to determine the risk of reading disability attributable to any single or combination of previously described reading disability genes. These studies are a crucial first step towards subsequent studies to try to translate these advances in our understanding of the genetics of reading disability to actual practical uses in populations, such as screening. An accurate and cost-effective population screening tool for reading disability would be useful for early detection of children at risk who would benefit from prevention programs, for resource planning by school districts, for detecting older children who are struggling in school in whom the diagnosis may have been missed or misdiagnosed, for adult literacy programs, and for tailoring prevention and intervention programs to specific students.